Long-term expression is essential for virtually all potential gene therapy treatments of neurodegenerative disorders. Some progress has been reported on obtaining long-term expression: Lentivirus vectors have been shown to support expression for 8 months, adeno-associated virus vectors for 6 months, adenovims vectors for 4 months, and helper virus-free HSV-1 vectors for 6 months. Thus, the current levels of long-term expression require significant improvement for human gene therapy applications in the brain. Herpes Simplex Virus (HSV-1) vectors are attractive for gene therapy of neurological disorders because HSV- can persist indefinitely in neurons, because large HSV-1 vectors can co express multiple genes, and because the approximate 150 kb genome of an HSV-1 vector can be viewed as a minichromosome. This laboratory has developed a helper virus-free HSV-1 plasmid vector system for gene transfer into neurons. We have demonstrated long-term (1 year) biochemical and behavioral correction of the rat model of Parkinson's disease by delivery of a HSV-1 vector that expresses human tyrosine hydroxylase (TH) into the partially denervated striatum. The two goals of this proposal are 1) to develop HSV-1 vectors that support significantly higher levels of long-term expression than have been achieved to date using any virus vector system, and 2) to establish general, mechanistic principles for enhancing long-term expression. Our approach is to view the large, approximate 150 kb genome of an HSV-1 vector as a minichromosome. We will use boundary elements to create a euchromatin-like domain, and we will use enhancers to turn on expression from specific promoters located within the euchromatin-like domain. In support of this approach, we have improved long-term expression by inserting one type of boundary element, an insulator, between the vector backbone and a transcription unit. Furthermore, we have shown that addition of an enhancer from the TH promoter to a neurofilament promoter supports long-term (6 months) expression. The first specific aim will isolate specific boundary elements and enhancers that support long-term expression. The second specific aim will investigate the mechanisms by which these elements support tong-term expression. The third specific aim will use these boundary elements and enhancers in combination to construct preferred vectors for long-term expression. Vectors will be evaluated for long-term expression in the rat striatum.